Merge 4b1c830957 into c5098961b0

[Youtube] Rework n function extraction pattern
Now also succeeds with player b12cc44b
2024-12-18 14:41:59 +00:00 · 2024-08-21 22:32:50 -04:00 · 2024-08-06 20:59:09 +01:00 · 2024-08-06 20:51:38 +01:00 · 2024-08-01 19:18:34 +01:00 · 2023-02-02 16:37:30 -08:00
8 changed files with 224 additions and 32 deletions
--- a/test/test_jsinterp.py
+++ b/test/test_jsinterp.py
@ -425,6 +425,34 @@ class TestJSInterpreter(unittest.TestCase):
            self._test(jsi, [''], args=['', '-'])
            self._test(jsi, [], args=['', ''])

+    def test_slice(self):
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice()}', [0, 1, 2, 3, 4, 5, 6, 7, 8])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(0)}', [0, 1, 2, 3, 4, 5, 6, 7, 8])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(5)}', [5, 6, 7, 8])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(99)}', [])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(-2)}', [7, 8])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(-99)}', [0, 1, 2, 3, 4, 5, 6, 7, 8])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(0, 0)}', [])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(1, 0)}', [])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(0, 1)}', [0])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(3, 6)}', [3, 4, 5])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(1, -1)}', [1, 2, 3, 4, 5, 6, 7])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(-1, 1)}', [])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(-3, -1)}', [6, 7])
+        self._test('function f(){return "012345678".slice()}', '012345678')
+        self._test('function f(){return "012345678".slice(0)}', '012345678')
+        self._test('function f(){return "012345678".slice(5)}', '5678')
+        self._test('function f(){return "012345678".slice(99)}', '')
+        self._test('function f(){return "012345678".slice(-2)}', '78')
+        self._test('function f(){return "012345678".slice(-99)}', '012345678')
+        self._test('function f(){return "012345678".slice(0, 0)}', '')
+        self._test('function f(){return "012345678".slice(1, 0)}', '')
+        self._test('function f(){return "012345678".slice(0, 1)}', '0')
+        self._test('function f(){return "012345678".slice(3, 6)}', '345')
+        self._test('function f(){return "012345678".slice(1, -1)}', '1234567')
+        self._test('function f(){return "012345678".slice(-1, 1)}', '')
+        self._test('function f(){return "012345678".slice(-3, -1)}', '67')
+

 if __name__ == '__main__':
    unittest.main()
--- a/test/test_youtube_signature.py
+++ b/test/test_youtube_signature.py
@ -174,6 +174,14 @@ _NSIG_TESTS = [
        'https://www.youtube.com/s/player/5604538d/player_ias.vflset/en_US/base.js',
        '7X-he4jjvMx7BCX', 'sViSydX8IHtdWA',
    ),
+    (
+        'https://www.youtube.com/s/player/20dfca59/player_ias.vflset/en_US/base.js',
+        '-fLCxedkAk4LUTK2', 'O8kfRq1y1eyHGw',
+    ),
+    (
+        'https://www.youtube.com/s/player/b12cc44b/player_ias.vflset/en_US/base.js',
+        'keLa5R2U00sR9SQK', 'N1OGyujjEwMnLw',
+    ),
 ]


--- a/youtube_dl/extractor/extractors.py
+++ b/youtube_dl/extractor/extractors.py
@ -1069,7 +1069,7 @@ from .rtve import RTVEALaCartaIE, RTVELiveIE, RTVEInfantilIE, RTVELiveIE, RTVETe
 from .rtvnh import RTVNHIE
 from .rtvs import RTVSIE
 from .ruhd import RUHDIE
-from .rumble import RumbleEmbedIE
+from .rumble import RumbleIE
 from .rutube import (
    RutubeIE,
    RutubeChannelIE,
@ -1439,6 +1439,7 @@ from .urort import UrortIE
 from .urplay import URPlayIE
 from .usanetwork import USANetworkIE
 from .usatoday import USATodayIE
+from .usawatchdog import UsaWatchdogIE
 from .ustream import UstreamIE, UstreamChannelIE
 from .ustudio import (
    UstudioIE,
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@ -132,6 +132,7 @@ from .kinja import KinjaEmbedIE
 from .arcpublishing import ArcPublishingIE
 from .medialaan import MedialaanIE
 from .simplecast import SimplecastIE
+from .rumble import RumbleIE


 class GenericIE(InfoExtractor):
@ -3518,6 +3519,10 @@ class GenericIE(InfoExtractor):
            return self.playlist_from_matches(
                zype_urls, video_id, video_title, ie=ZypeIE.ie_key())

+        rumble_urls = RumbleIE.rumble_embedded_id(webpage)
+        if rumble_urls is not None:
+            return self.playlist_result(rumble_urls) if len(rumble_urls) > 1 else rumble_urls[0]
+
        # Look for HTML5 media
        entries = self._parse_html5_media_entries(url, webpage, video_id, m3u8_id='hls')
        if entries:
--- a/youtube_dl/extractor/rumble.py
+++ b/youtube_dl/extractor/rumble.py
@ -1,5 +1,6 @@
 # coding: utf-8
 from __future__ import unicode_literals
+import re

 from .common import InfoExtractor
 from ..compat import compat_str
@ -8,31 +9,100 @@ from ..utils import (
    int_or_none,
    parse_iso8601,
    try_get,
+    ExtractorError,
 )


-class RumbleEmbedIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?rumble\.com/embed/(?:[0-9a-z]+\.)?(?P<id>[0-9a-z]+)'
-    _TESTS = [{
-        'url': 'https://rumble.com/embed/v5pv5f',
-        'md5': '36a18a049856720189f30977ccbb2c34',
-        'info_dict': {
-            'id': 'v5pv5f',
-            'ext': 'mp4',
-            'title': 'WMAR 2 News Latest Headlines | October 20, 6pm',
-            'timestamp': 1571611968,
-            'upload_date': '20191020',
-        }
-    }, {
-        'url': 'https://rumble.com/embed/ufe9n.v5pv5f',
-        'only_matching': True,
-    }]
+class RumbleIE(InfoExtractor):

-    def _real_extract(self, url):
-        video_id = self._match_id(url)
+    RE_DICT = {
+        'iframe_url': {
+            're': r'https?://(?:www\.)?rumble\.com/embed/(?:[0-9a-z]+\.)?(?P<id>[0-9a-z]+)',
+            'compiled': None},
+        'jscript_url': {
+            're': r'https?://rumble\.com/[a-zA-Z0-9-_.]*\.html',
+            'compiled': None},
+        'list_url': {
+            're': r'https?://rumble.com/(?:c|user)/(?P<id>[^/]+)',
+            'compiled': None},
+        'jscript_id': {
+            're': r'Rumble *\( *["\']play["\'], *\{[^}]*["\']video["\'] *: *["\'](?P<id>[^"\']+)',
+            'compiled': None}
+    }
+
+    _TESTS = [
+        {
+            'url': 'https://rumble.com/embed/v5pv5f',
+            'md5': '36a18a049856720189f30977ccbb2c34',
+            'info_dict': {
+                'id': 'v5pv5f',
+                'ext': 'mp4',
+                'title': 'WMAR 2 News Latest Headlines | October 20, 6pm',
+                'timestamp': 1571611968,
+                'upload_date': '20191020',
+            }
+        },
+        {
+            'url': 'https://rumble.com/v8c1bt-wmar-2-news-latest-headlines-october-20-6pm.html',
+            'md5': '36a18a049856720189f30977ccbb2c34',
+            'info_dict': {
+                'id': 'v5pv5f',
+                'ext': 'mp4',
+                'title': 'WMAR 2 News Latest Headlines | October 20, 6pm',
+                'timestamp': 1571611968,
+                'upload_date': '20191020',
+            }
+        },
+        {
+            'url': 'https://rumble.com/c/PeakProsperity',
+            'playlist_mincount': 25,
+            'info_dict': {
+                'id': 'PeakProsperity',
+            }
+        },
+        {
+            'url': 'https://rumble.com/embed/ufe9n.v5pv5f',
+            'only_matching': True,
+        }
+    ]
+
+    @classmethod
+    def get_re(cls, tag):
+        if cls.RE_DICT[tag]['compiled'] is None:
+            cls.RE_DICT[tag]['compiled'] = re.compile(cls.RE_DICT[tag]['re'])
+        return cls.RE_DICT[tag]['compiled']
+
+    @classmethod
+    def suitable(cls, url):
+        return (cls.get_re('jscript_url').match(url) is not None
+                or cls.get_re('list_url').match(url) is not None
+                or cls.get_re('iframe_url').match(url) is not None)
+
+    @staticmethod
+    def rumble_embedded_id(page_data):
+        '''For use by extractors of sites which use emedded Rumble videos. Given
+        a webpage as a string returns a list of url result dicts for each embedded
+        rumble video found. None is returned if no embeds were found. Duplicates
+        are not removed'''
+
+        embeds = []
+        # The JS embeds
+        for mobj in RumbleIE.get_re('jscript_id').finditer(page_data):
+            embeds.append(InfoExtractor.url_result('https://rumble.com/embed/' + mobj.group('id'), 'Rumble', mobj.group('id')))
+
+        # The iframes embeds
+        for mobj in RumbleIE.get_re('iframe_url').finditer(page_data):
+            embeds.append(InfoExtractor.url_result('https://rumble.com/embed/' + mobj.group('id'), 'Rumble', mobj.group('id')))
+
+        return embeds if embeds else None
+
+    def rumble_video_info(self, video_id):
        video = self._download_json(
            'https://rumble.com/embedJS/', video_id,
            query={'request': 'video', 'v': video_id})
+        if not video:
+            raise ExtractorError('Unable to locate video information.', expected=True)
+
        title = video['title']

        formats = []
@ -65,3 +135,23 @@ class RumbleEmbedIE(InfoExtractor):
            'channel_url': author.get('url'),
            'duration': int_or_none(video.get('duration')),
        }
+
+    def _real_extract(self, url):
+        if self.get_re('jscript_url').match(url) is not None:
+            page = self._download_webpage(url, 'Rumble Page')
+            video_id = self._search_regex(self.get_re('jscript_id'), page, "id")
+            return self.rumble_video_info(video_id)
+
+        mobj = self.get_re('list_url').match(url)
+        if mobj is not None:
+            urls = []
+            id = mobj.group('id')
+            page = self._download_webpage(url, id)
+            for mobj in re.finditer(r'<a class=video-item--a href=\/(?P<href>[a-zA-Z0-9\-.]+)>', page):
+                urls.append('https://rumble.com/' + mobj.group('href'))
+
+            return self.playlist_from_matches(urls, id)
+
+        mobj = self.get_re('iframe_url').match(url)
+        if mobj is not None:
+            return self.rumble_video_info(mobj.group('id'))
--- a/youtube_dl/extractor/usawatchdog.py
+++ b/youtube_dl/extractor/usawatchdog.py
@ -0,0 +1,24 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+import re
+
+from .common import InfoExtractor
+
+
+class UsaWatchdogIE(InfoExtractor):
+    _VALID_URL = r'^https?://(?:www\.)?usawatchdog\.com/?$'
+    _TEST = {
+        'url': 'https://usawatchdog.com/',
+        'playlist_mincount': 15,
+        'info_dict': {
+            'id': 'USA Watchdog',
+        }}
+
+    def _real_extract(self, url):
+        matches = []
+        for mobj in re.finditer(r'front-view-title[^<]+<a.+href=["\'](?P<href>https?:(?:www\.)?//usawatchdog.com/[^/]+\/?)[^>]+>(?P<title>[^<]+)',
+                                self._download_webpage(url, 'Site Root')):
+            matches.append(self.url_result(mobj.group('href'),
+                                           video_title=mobj.group('title').encode('utf8')))
+
+        return self.playlist_result(matches, 'USA Watchdog')
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@ -1659,17 +1659,46 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
    def _extract_n_function_name(self, jscode):
        func_name, idx = self._search_regex(
            # new: (b=String.fromCharCode(110),c=a.get(b))&&c=nfunc[idx](c)
-            # or:  (b="nn"[+a.D],c=a.get(b))&&(c=nfunc[idx](c)s
-            # old: .get("n"))&&(b=nfunc[idx](b)
-            # older: .get("n"))&&(b=nfunc(b)
+            # or:  (b="nn"[+a.D],c=a.get(b))&&(c=nfunc[idx](c)
+            # or:  (PL(a),b=a.j.n||null)&&(b=nfunc[idx](b)
+            # or:  (b="nn"[+a.D],vL(a),c=a.j[b]||null)&&(c=narray[idx](c),a.set(b,c),narray.length||nfunc("")
+            # old: (b=a.get("n"))&&(b=nfunc[idx](b)(?P<c>[a-z])\s*=\s*[a-z]\s*
+            # older: (b=a.get("n"))&&(b=nfunc(b)
            r'''(?x)
-                (?:\(\s*(?P<b>[a-z])\s*=\s*(?:
-                    String\s*\.\s*fromCharCode\s*\(\s*110\s*\)|
-                    "n+"\[\s*\+?s*[\w$.]+\s*]
-                )\s*,(?P<c>[a-z])\s*=\s*[a-z]\s*)?
-                \.\s*get\s*\(\s*(?(b)(?P=b)|"n{1,2}")(?:\s*\)){2}\s*&&\s*\(\s*(?(c)(?P=c)|b)\s*=\s*
-                (?P<nfunc>[a-zA-Z_$][\w$]*)(?:\s*\[(?P<idx>\d+)\])?\s*\(\s*[\w$]+\s*\)
-            ''', jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
+                \((?:[\w$()\s]+,)*?\s*      # (
+                (?P<b>[a-z])\s*=\s*         # b=
+                (?:
+                    (?:                     # expect ,c=a.get(b) (etc)
+                        String\s*\.\s*fromCharCode\s*\(\s*110\s*\)|
+                        "n+"\[\s*\+?s*[\w$.]+\s*]
+                    )\s*(?:,[\w$()\s]+(?=,))*|
+                       (?P<old>[\w$]+)      # a (old[er])
+                   )\s*
+                   (?(old)
+                                            # b.get("n")
+                       (?:\.\s*[\w$]+\s*|\[\s*[\w$]+\s*]\s*)*?
+                       (?:\.\s*n|\[\s*"n"\s*]|\.\s*get\s*\(\s*"n"\s*\))
+                       |                    # ,c=a.get(b)
+                       ,\s*(?P<c>[a-z])\s*=\s*[a-z]\s*
+                       (?:\.\s*[\w$]+\s*|\[\s*[\w$]+\s*]\s*)*?
+                       (?:\[\s*(?P=b)\s*]|\.\s*get\s*\(\s*(?P=b)\s*\))
+                   )
+                                            # interstitial junk
+                   \s*(?:\|\|\s*null\s*)?(?:\)\s*)?&&\s*(?:\(\s*)?
+               (?(c)(?P=c)|(?P=b))\s*=\s*   # [c|b]=
+                                            # nfunc|nfunc[idx]
+                   (?P<nfunc>[a-zA-Z_$][\w$]*)(?:\s*\[(?P<idx>\d+)\])?\s*\(\s*[\w$]+\s*\)
+            ''', jscode, 'Initial JS player n function name', group=('nfunc', 'idx'),
+            default=(None, None))
+        # thx bashonly: yt-dlp/yt-dlp/pull/10611
+        if not func_name:
+            self.report_warning('Falling back to generic n function search')
+            return self._search_regex(
+                r'''(?xs)
+                    (?:(?<=[^\w$])|^)       # instead of \b, which ignores $
+                    (?P<name>(?!\d)[a-zA-Z\d_$]+)\s*=\s*function\((?!\d)[a-zA-Z\d_$]+\)
+                    \s*\{(?:(?!};).)+?["']enhanced_except_
+                ''', jscode, 'Initial JS player n function name', group='name')
        if not idx:
            return func_name

--- a/youtube_dl/jsinterp.py
+++ b/youtube_dl/jsinterp.py
@ -925,9 +925,16 @@ class JSInterpreter(object):
                    obj.reverse()
                    return obj
                elif member == 'slice':
-                    assertion(isinstance(obj, list), 'must be applied on a list')
-                    assertion(len(argvals) == 1, 'takes exactly one argument')
-                    return obj[argvals[0]:]
+                    assertion(isinstance(obj, (list, compat_str)), 'must be applied on a list or string')
+                    # From [1]:
+                    # .slice() - like [:]
+                    # .slice(n) - like [n:] (not [slice(n)]
+                    # .slice(m, n) - like [m:n] or [slice(m, n)]
+                    # [1] https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Array/slice
+                    assertion(len(argvals) <= 2, 'takes between 0 and 2 arguments')
+                    if len(argvals) < 2:
+                        argvals += (None,)
+                    return obj[slice(*argvals)]
                elif member == 'splice':
                    assertion(isinstance(obj, list), 'must be applied on a list')
                    assertion(argvals, 'takes one or more arguments')
Author	SHA1	Message	Date
epilitimus	dd261d7b9f	Merge `4b1c830957` into `c5098961b0`	2024-08-21 22:32:50 -04:00
dirkf	c5098961b0	[Youtube] Rework n function extraction pattern Now also succeeds with player b12cc44b	2024-08-06 20:59:09 +01:00
dirkf	dbc08fba83	[jsinterp] Improve slice implementation for player b12cc44b Partly taken from yt-dlp/yt-dlp#10664, thx seproDev Fixes #32896	2024-08-06 20:51:38 +01:00
Aiur Adept	71223bff39	[Youtube] Fix nsig extraction for player 20dfca59 (#32891 ) * dirkf's patch for nsig extraction * add generic search per yt-dlp/yt-dlp/pull/10611 - thx bashonly --------- Co-authored-by: dirkf <fieldhouse@gmx.net>	2024-08-01 19:18:34 +01:00
Glenn Pavlovic	4b1c830957	fix formatting issue	2023-02-02 16:37:30 -08:00
Glenn Pavlovic	a777aeeda0	combine all the rumble extractors, add rumble to generic.py, remove UsaWatchdogStory	2023-01-29 18:20:46 -08:00
Glenn Pavlovic	7bb8d94184	Rumble + UsaWatchdog - improves Rumble support and adds UsaWatchdog support	2023-01-17 18:40:37 -08:00